Educational dialogue among teachers experiencing different levels of self-efficacy

This is a sel -a chi ed e sion o an o iginal a icle. This e sion may di e om he o iginal in pagina ion and ypog aphic de ails. Au ho (s): Ti le: Yea : Ve sion: Copy igh : Righ s: Righ s u l: Please ci e he o iginal e sion: CC BY-NC-ND 4.0 h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0/ Educa ional dialogue among eache s expe iencing di e en le els o sel -e icacy © 2021 Else ie L d. All igh s ese ed. Accep ed e sion (Final d a ) Muhonen, Heli; Paka inen, Eija; Rasku-Pu onen, Helena; Le kkanen, Ma ja- K is iina Muhonen, H., Paka inen, E., Rasku-Pu onen, H., & Le kkanen, M.-K. (2021). Educa ional dialogue among eache s expe iencing di e en le els o sel -e icacy. Lea ning, Cul u e and Social In e ac ion, 29, A icle 100493. h ps://doi.o g/10.1016/j.lcsi.2021.100493 2021 Educa ional dialogue among eache s expe iencing di e en le els o sel -e icacy Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 1 1. In oduc ion Educa ional dialogue e e s o ecip ocal, cumula i e and ex ended class oom alk, in which he eache and child en oge he pa icipa e in sha ed knowledge building wi h he aim o eaching educa ional lea ning goals (Alexande , 2008; Me ce & Li le on, 2007; Michaels & O'Conno , 2015). Dialogic p ac ices ha e been shown o c ea e mo i a ionally suppo i e lea ning en i onmen s o s uden s (Böheim e al., 2021) and suppo he ac ual s uden lea ning p ocess (e.g., Alexande , 2018; Muhonen e al., 2018; Howe e al., 2019; Sedo a e al., 2019). Despi e he bene i s o educa ional dialogue o s uden s’ lea ning, s udies ha e shown ha educa ional dialogue a ely occu s in he class ooms and ha eache s a ely use sca olding s a egies ha suppo educa ional discou se and s uden pa icipa ion in i (Howe & Abedin, 2013; Webb e al., 2009). The idealis ic app oach o dialogue does no seem o ind i s way o class ooms i sel , which is why he e is a need o ind ou which eache - ela ed ac o s may be connec ed wi h he occu ence and quali y o dialogue in he class ooms. The eache - ela ed ac o o sel -e icacy in eaching and lea ning has in e es ed esea che s o decades (Koomen & Zee, 2016). Teache s’ sel -e icacy has been associa ed wi h di e se domains o eache pe o mance. P io esea ch has shown ha eache s’ sel - e icacy belie s os e i al eache - ela ed ou comes, such as highe wo k sa is ac ion (G anzie a & Pe e a, 2019; Klassen & Chiu, 2010) and commi men (Klassen & Chiu, 2011). Teache s’ sel -e icacy also has an impac on he educa ional ac i i ies and eaching p ac ices ha eache s choose o use and in es e o and pe sis ence in (Bandu a, 1997; Wool olk Hoy e al., 2009). In ac , p e ious esea ch has shown ha high sel -e icacy eache s can c ea e a mo e a ou able ins uc ional lea ning en i onmen by u ilising, o ins ance, mo e s uden - cen ed ins uc ional app oaches (Nie e al., 2013) and p o iding g ea e lea ning suppo (Holzbe ge e al., 2013). Teache s’ sel -e icacy has also been ound o be posi i ely associa ed wi h he quali y o class oom in e ac ions in e ms o emo ional suppo , class oom Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 2 o ganisa ion, and ins uc ional suppo (Pe e a & John, 2020). Howe e , he e is a lack o s udies in es iga ing how he le el o eache sel -e icacy may be linked wi h he educa ional dialogue ha he eache and child en ha e in he class oom. The p esen s udy aims o con ibu e o his ield by in es iga ing how he occu ence and quali y o educa ional dialogue in class ooms e lec s he di e ing le els o eache sel -e icacy in a sample o p ima y school eache s. 1.1. Educa ional dialogue The idea o he impo ance o educa ional dialogue is based on Vygo sky’s sociocul u al heo y (1978), which emphasises he meaning o social in e ac ions o s uden lea ning and de elopmen , highligh ing he i al ole o language as bo h a cul u al media o and ool o hinking. In e ec i e educa ional dialogue, s uden s ha e he oppo uni y o pa icipa e ac i ely by sha ing hei hough s, elabo a ing on hei iews, and building on each o he ’s ideas (Alexande , 2008; Michaels & O'Conno , 2015; Wilkinson e al., 2017). Se e al concep s and de ini ions ha e been used o desc ibe educa ional dialogue in he class oom. The cons uing o educa ional dialogue can a y depending on he ocus o a en ion, whe he i is on a whole- class discussion be ween eache and s uden s o on pee dialogue. In addi ion, he educa ional dialogue can be cons ued di e en ly i he ocus is on he ac ions and sca olding o he eache , s uden in ol emen o he whole-class p ocess o dialogic exchanges. The p edominan ocus o he p esen s udy is on whole-class dialogue and he eache ’s sca olding wi hin i . The e o e, he p esen s udy elies on Alexande ’s (2008) concep o dialogic eaching, in which he educa ional dialogue be ween he eache and s uden s is de ined as exchanges whe e he pa icipan s ask ques ions, explain hei poin s o iew and make commen s abou each o he ’s ideas, all o which can p omp u he ques ions and elabo a ions. The concep o dialogic eaching includes i e p inciples (2018, p. 566) ha desc ibe and ame Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 3 educa ional dialogue, as ollows: ‘1) collec i e ( he class oom is a si e o join lea ning and enqui y); 2) ecip ocal (pa icipan s lis en o each o he , sha e ideas and conside al e na i e iewpoin s); 3) suppo i e (pa icipan s eel able o exp ess ideas eely, wi hou he isk o emba assmen o e ‘w ong’ answe s); 4) cumula i e (pa icipan s build on hei own and each o he ’s con ibu ions and chain hem in o cohe en lines o hinking and unde s anding); and 5) pu pose ul (class oom alk, hough open and dialogic, is s uc u ed wi h speci ic lea ning goals in iew)’. In addi ion o he i e p inciples om Alexande , Le s ein (2006) has sugges ed wo mo e c i e ia as impo an dialogic ea u es: dialogue should be c i ical ( eache and s uden s iden i y and in es iga e poin s and explo e ques ions oge he ) and meaning ul ( eache s and s uden s ela e o he discussion opic and exp ess hei pe sonal iews). Fo educa ional dialogue o occu , he eache mus play a i al ole in enabling oppo uni ies o s uden pa icipa ion and concep ual de elopmen h ough sca olding, which includes s a egies such as inqui y, open ques ions, eedback, nego ia ing consensus and assis ing s uden s in explaining hei own hinking (Gillies, 2004; Gillies, 2015; Rojas- D ummond e al., 2013). In well-sca olded educa ional dialogue, sha ed knowledge building is s uc u ed h ough ques ioning, join ac i i y and sha ed concep ions (Alexande , 2000; Rojas-D ummond e al., 2013). Al hough i is he eache s who ypically ini ia e and manage he educa ional dialogues (Wells, 2009), hey should also be sensi i e o s uden s’ ini ia i es by, o example, allowing space o new u ns in he discussion based on s uden s’ commen s, asking au hen ic ques ions ela ed o he s uden s’ li es and including s uden s’ esponses in o he ques ions ha ollow (Nys and, 1997). Acknowledging hese child-cen ed sca olding s a egies is o high impo ance because s uden s a ely pa icipa e and make ini ia i es in educa ional dialogue wi hou he eache ’s speci ic encou agemen (Chinn e al., 2000; Myhill, 2006). Ex ending he wo k o Rasku-Pu onen e al. (2012), Muhonen e al. (2016) explo ed he di e en ypes o educa ional dialogues ega ding he ini ia o (ei he he eache o s uden ) Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 4 and he quali y (ei he mode a e o high) o he dialogue and we e able o iden i y ou pa e ns o dialogic eaching: eache -ini ia ed dialogues o a mode a e and high quali y and s uden - ini ia ed dialogues o a mode a e and high quali y. In he pa e ns, he quali y o he dialogue was ound o di e based on he eache ’s sca olding and he s uden s’ pa icipa ion. These ou pa e ns will also be u ilised in he p esen s udy. As shown, he esea ch on educa ional class oom alk has ocused on he quali y, acili a ing s uc u es and ou comes o p oduc i e educa ional dialogues be ween he eache and s uden s (Li le on & Howe, 2010). Howe e , less is known abou he eache - ela ed ac o s ha may link wi h he eache ’s abili y o conduc educa ional dialogue in he class oom. A ecen s udy by Muhonen e al. (2020) in es iga ed he eache s’ le els o s ess on he backg ound o educa ional dialogue, inding ha mode a ely s essed eache s conduc ed educa ional dialogue he mos , whe eas eache s wi h low s ess conduc ed dialogue he leas . In he p esen s udy, he eache - ela ed ac o o sel -e icacy is conside ed when analysing he occu ence and quali y o educa ional dialogue. 1.2. Teache sel -e icacy The cons uc o eache sel -e icacy is la gely g ounded in he social cogni i e heo y o Bandu a (1986), who sugges ed ha people who belie e ha hey will be success ul in he gi en ask a e mo e likely o achie e he desi ed esul s; hese people in es subs an ial e o , s ay pe sis en in he ace o se backs and de elop coping mechanisms o managing he se backs. E icacy belie s a e no only he gene al judgemen s o a pe son’s skills and abili ies bu belie s o wha he pe son can do wi h wha e e skills and abili ies he pe son possesses (Bong & Skaal ik, 2003). In he ield o educa ion, eache sel -e icacy is de ined as indi idual eache belie s abou hei own abili ies o plan, o ganise, and ca y ou he ac i i ies equi ed o a ain gi en educa ional goals (Skaal ik & Skaal ik, 2007; Tschannen-Mo an e al., 1998). Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 5 This de ini ion o eache sel -e icacy includes bo h eache s’ sel -pe cep ions o pe sonal eaching capabili y and he judgemen s abou he equi emen s o speci ic eaching asks (including he assessmen o ex e nal cons ain s and esou ces) (Pe e a e al., 2019). P io esea ch has shown ha eache s’ posi i e sel -e icacy belie s a e linked o hei high men al wellbeing, wo k engagemen and job sa is ac ion (see Zee & Koomen, 2016 o a e iew). In addi ion, hei sense o e icacy ends o inc ease wi h he yea s o expe ience in eaching (Soodak & Podell, 1997). On he o he hand, eaching expe ience may also ha e an e ec on how eache s pe cei e hei own abili ies and limi a ions and how hey become mo e awa e o hem (Egyed & Sho , 2006). Teache s wi h a highe sense o e icacy a e mo e likely o s ay in eaching ca ee s (Glickman & Tamashi o, 1982), and he ex ended eaching expe ience can inc ease eache s’ epe oi e o e ec i e eaching echniques (Egyed & Sho , 2006). Based on he social cogni i e p ocess heo y o eache sel -e icacy (Tschannen-Mo an & Wool olk Hoy, 2001; Wool olk Hoy e al., 2009; Zee & Koomen, 2016), eache s wi h high sel -e icacy a e expec ed o include a highe quali y o eache –s uden in e ac ions in hei class ooms (Pe e a & John, 2020). Teache s who expe ience a s ong sense o sel -e icacy a e ound o be mo e ocused on planning and accomplishing eaching- ela ed goals (Skaal ik & Skaal ik, 2007; Tschannen-Mo an & Wool olk Hoy, 2001), which can be e lec ed in hei class oom beha iou and in e ac ions. Fo example, e icacious eache s end o u ilise mo e s uden -cen ed eaching s a egies o suppo s uden s’ engagemen and lea ning (Küns ing e al., 2016; Nie e al., 2013) and a e mo e e ec i e in p o iding eedback on s uden pe o mance and aking he ime o co ec possible misunde s andings (Ash on & Webb, 1986; Gibson & Dembo, 1984). In addi ion, eache s wi h high sel -e icacy ha e been shown o conduc high- quali y class oom o ganisa ion (Küns ing e al., 2016, Tsouloupas e al., 2010). Based on hese indings, i can be expec ed ha highe eache sel -e icacy is mo e likely o lead o highe Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 6 quali y in he eache –s uden in e ac ion (Wool olk Hoy e al., 2009). Suppo ing his assump ion, a ecen s udy by Pe e a and John (2020) showed ha eache s’ sel -e icacy belie s o eaching ma h we e posi i ely associa ed wi h class oom in e ac ion quali y as measu ed by he domains o emo ional suppo , class oom o ganisa ion and ins uc ional suppo . Al hough p e ious esea ch has been able o show he impo ance o eache sel -e icacy o eache –s uden in e ac ions a a mo e gene al class oom in e ac ion le el, he e a e no empi ical s udies ha ha e in es iga ed he ole o eache sel -e icacy in he ac ual educa ional dialogue be ween eache and s uden s. Because educa ional dialogue can be conside ed a o m o high-quali y in e ac ion be ween eache s and s uden s, we may expec highe eache sel -e icacy o link wi h he educa ional dialogue in he class ooms. 1.3. The p esen s udy The e is e y li le p io esea ch a ailable on he link be ween eache - ela ed ac o s and educa ional dialogue ha can help in unde s anding he condi ions ha may a ou o es ic educa ional dialogue and s uden s’ lea ning h ough i . In he p esen s udy, he eache - ela ed ac o o sel -e icacy is conside ed, he e explo ing he occu ence and quali y o educa ional dialogue among class ooms o eache s expe iencing di e en le els o s ess. Because o he quali a i e na u e o he p esen s udy and lack o p e ious s udies on he ela ion be ween eache sel -e icacy and educa ional dialogue, we a e no able o se a clea hypo hesis o he s udy. Howe e , based on he indings in p e ious esea ch on eache sel - e icacy belie s and class oom in e ac ion, we can expec o ind mo e and highe -quali y educa ional dialogue in he class ooms o eache s wi h high sel -e icacy. The p esen s udy in es iga es he ollowing esea ch ques ions: Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 7 1. How many episodes o educa ional dialogue can be ound wi hin he class ooms o eache s wi h low, mode a e and high sel -e icacy? 2. Wha kinds o pa e ns o dialogic eaching can be iden i ied among eache s wi h di e en sel -e icacy le els, and how does he occu ence o hese pa e ns a y among eache s? 2. Me hod 2.1. Pa icipan s and p ocedu e The p esen s udy is pa o a la ge longi udinal esea ch p ojec ocusing on he in e ac ion and wellbeing o eache s and s uden s (Le kkanen & Paka inen, 2016–2022). The pa icipan s o he p esen s udy we e Finnish G ade 1 eache s and he s uden s in hei class ooms. On a e age, he eache s (50 emale, 4 male) we e 44.6 yea s old and holde s o a mas e ’s deg ee, which is equi ed o p ima y school eache s in Finland. Du ing he da a collec ion, he e we e, on a e age, 17.8 s uden s p esen in he class ooms (minimum 6, maximum 23). This numbe o s uden s e lec s he ypical a e age class sizes in Finnish G ade 1 class ooms. The child en (n = 780) we e app oxima ely se en yea s old, 49% o hem being gi ls and 51% being boys. The child en’s pa en s (n = 577) epo ed hei educa ional deg ee, which a ied om no oca ional educa ion o a licen ia e o doc o a e (Mode = oca ional school deg ee). The pa icipa ing eache s and hei pa en s ga e hei w i en consen o hei o hei child’s pa icipa ion. The pa icipan s we e ec ui ed on a olun a y basis, and hey we e awa e o he possibili y o d opping ou a any poin . The esea ch p ojec ecei ed e hical app o al om he uni e si y’s e hics commi ee in 2017 be o e he p esen da a we e collec ed. The eache s illed ou a ques ionnai e ega ding hei wo k- ela ed sel -e icacy belie s and eaching expe ience. E e y class oom was also ideo eco ded du ing one school day, wi h h ee o ou lessons ha ypically las ed 45 minu es. The eco dings we e conduc ed wi h wo Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 14 3.2. Quali y o educa ional dialogue: Pa e ns o dialogic eaching Based on he second esea ch ques ion, Table 1 ep esen s he equencies o how he 105 iden i ied episodes o educa ional dialogue we e ca ego ised as ep esen ing he pa e ns o dialogic eaching among he low, mode a e and high sel -e icacy eache s. Teache -ini ia ed dialogue o a high quali y was he p edominan dialogue ype (n = 59) in he sample. Howe e , he K uskal–Wallis H es showed ha he e was a s a is ically signi ican di e ence in he numbe o dialogic episodes ep esen ing he pa e n among he h ee g oups o eache s, χ2(2) = 8.00, p = 0.018, wi h a mean ank sco e o 7.00 o eache s wi h low sel -e icacy, 16.44 o eache s wi h mode a e sel -e icacy and 14.06 o eache s wi h high sel -e icacy. The e was also a ma ginally signi ican di e ence in he numbe o dialogic episodes ep esen ing he s uden -ini ia ed dialogue o a high quali y pa e n among he h ee g oups o eache s, χ2(2) = 5.36, p = 0.069, wi h a mean ank sco e o 11.63 o eache s wi h low sel -e icacy, 9.44 o eache s wi h mode a e sel -e icacy and 16.44 o eache s wi h high sel -e icacy. Wi hin he pa e ns o eache -ini ia ed dialogue o mode a e quali y (n = 25) and child-ini ia ed dialogue o mode a e quali y (n = 6) no signi ican di e ences among he h ee g oups o eache s we e ound. The nex h ee sec ions desc ibe in mo e de ail he educa ional dialogues ound wi hin he class ooms o each eache g oup. 3.2.1. Educa ional dialogue o low sel -e icacy eache s Low sel -e icacy eache s we e ound o u ilise educa ional dialogue he leas in he sample. Compa ed wi h bo h mode a e sel -e icacy eache s (z = -7.437, p =.034), and high sel -e icacy eache s (z = -7.562, p =.031), he pos -hoc es using Dunn’s es showed ha he In o al 25 59 6 15 105 K uskal–Wallis es x2 (p- alue) .220 (.896) 8.000 (.018) 1.278 (.528) 5.36 (.069) 6.116 (.047) Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 15 equency di e ences in he o al numbe o dialogic episodes we e s a is ically signi ican . Ou o he 18 episodes o educa ional dialogue iden i ied among he low sel -e icacy eache s, eache -ini ia ed dialogue o mode a e quali y was he mos common pa e n (n = 8) (see Table 1). The pa e ns o eache -ini ia ed dialogue o high quali y n = 5), child-ini ia ed dialogue o mode a e quali y (n = 2) and child-ini ia ed dialogue o high quali y (n = 3) occu ed less o en in he class ooms o low sel -e icacy eache s. Example 1 p esen s an ex ac o a eache -ini ia ed mode a e-quali y dialogue, which was he mos common pa e n o dialogic eaching ound in he class ooms o he low sel - e icacy eache s. The dialogue was iden i ied du ing a li e acy lesson. The ela i ely b ie dialogue is buil a ound a ques ion o which he eache is looking o a ce ain speci ic explana ion (Do you know wha eading wi h exp essions means?). The child en pa icipa e in he discussion by sugges ing answe s o he eache ’s ques ions (A bed ime s o y! ; So ha you y o speak and ead wi h a co ec oice.). The eache is suppo i e o he s uden s’ answe s and p o ides con i ming and expanding eedback (Well yes, i can be a bed ime s o y also. ; Tha ’s igh . Fo example, he e in his pa , when Annie says ‘The escue is he e’ she p obably does no only say i no mally bu e y is e y exci edly ‘The escue is he e!!!). Howe e , in gene al, he eache ’s sca olding is ela i ely limi ed. Fo ins ance, he discussion does no include any cla i ying o ex ending ques ions om he eache , which could ha e b oadened he dialogue and encou age u he elabo a ion. Example 1. Teache -ini ia ed mode a e-quali y dialogue o a low sel -e icacy eache . Con ex : The eache and child en ha e been eading a homewo k chap e . Teache : Do you know wha eading wi h exp essions means? Could someone ell wha i means? Child 1: A bed ime s o y! Teache : Well yes i can be a bed ime s o y also. Bu does i mean when you ead i wi h exp essions? Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 16 Child 2: So ha you y o speak and ead wi h a co ec oice. Teache : Tha ’s igh . Fo example, he e in his pa , when Annie says, ‘The escue is he e’ she p obably does no only say i no mally bu e y exci edly, ‘The escue is he e!!!’ Child en (exci edly): ‘The escue is he e!!!’ Teache : Yes jus like ha . I know you can ead i like ha . Child 3: Yes I always y o ead like ha . Wi h ha di e en oice. Teache : Good. Would you like o y one mo e ime? The eache and child en con inue eading wi h exp ession 3.2.2. Educa ional dialogue o mode a e sel -e icacy eache s Mode a e sel -e icacy eache s we e ound o u ilise educa ional dialogue he mos (44 dialogic episodes, see Table 1). The majo i y o he dialogic episodes ep esen ed eache - ini ia ed dialogue o high quali y (30 episodes ou o he o al o 44). The pos -hoc es using Dunn’s es showed ha he mode a e sel -e icacy eache s u ilised signi ican ly mo e eache - ini ia ed dialogue o high quali y compa ed wi h he low sel -e icacy eache s (z = -9.437, p =.007). The o he ypes o pa e ns occu ed less o en in he class ooms o mode a e sel - e icacy eache s. In pa icula , he child-ini ia ed dialogues we e a e in he mode a e sel - e icacy eache sample: child-ini ia ed dialogue o mode a e quali y (n = 3) and child-ini ia ed dialogue o a high quali y (n = 2). In addi ion, nine episodes o eache -ini ia ed dialogue o mode a e quali y we e ound. Example 2 p esen s an ex ac o a eache -ini ia ed high-quali y dialogue, which is he p edominan pa e n o dialogic eaching ound in he class ooms o he mode a e sel -e icacy eache s. The dialogue was iden i ied du ing a ma hema ics lesson. In he dialogue, bo h he eache and child en con ibu e ac i ely o he discussion and sha ed knowledge-building p ocess. The eache ini ia es he discussion by i s asking a sho closed ques ion (So i ze o means none hen is he numbe hen comple ely useless?); a e , he s uden s’ esponse expands his by asking open ques ions (So why do we need ze o hen? ; Think abou i , why do you need Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 17 ze o and in which si ua ion ha e you used ze o? ). In his way, he eache encou aged s uden s o explain hei opinions and hinking. The s uden s pa icipa e ac i ely in he dialogue by sha ing hei ac ual knowledge wi h he eache ’s ques ions (In phone numbe s. ; In codes. ; So in his yea 2017 so he e is ze o in i .), bu hey also sha e hei opinions and iews, e en wi hou he eache ’s speci ic encou agemen (Yes, we do! Fo so many hings! ; I makes a big di e ence! ; Ac ually, we need ze o o e e y hing.). The eache p o ides encou aging eedback o he s uden s’ answe s (e.g., G ea obse a ion! The e in 2017 he e is a ze o; Yes, we need i . Ze o is ac ually a e y use ul numbe .) and expand on some o he commen s (Tha ’s igh ! We can explo e ha a bi mo e. … And i I add his (adds ze o a e one), how many is i hen?). Example 2. Teache -ini ia ed high-quali y dialogue o mode a e sel -e icacy eache . Con ex : The heme o he lesson is he numbe ze o. Teache : So i ze o means none hen is he numbe hen comple ely useless? Child en: No! Teache : So why do we need ze o hen? Child 1: We don’ need i o any hing. Teache : Don’ we? Child 2: Yes we do! Fo so many hings! Teache : Think abou i , why do you need ze o and in which si ua ion ha e you used ze o? Child 3: In phone numbe s. Teache : T ue. In phone numbe s, he e a e o en ze os. Child 4: In codes. Teache : Yes he e a e ze os in di e en codes. Child 5: So in his yea 2017 so he e is ze o in i . Teache : G ea obse a ion! The e in 2017 he e is a ze o. Child 6: In hund eds and ens, he e a e ze os. Teache Tha ’s igh ! We can explo e ha a bi mo e. Please look a he e (goes o he black boa d). Could you show me wi h you inge how many is his (d aws numbe 1). And i I add his (adds ze o a e one), how many is i hen? Child en: Ten!!! Child 7: I makes a big di e ence! Child 2: Ac ually we need ze o o e e y hing. Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 18 Teache : Yes, we need i . Ze o is ac ually a e y use ul numbe . Teache child en con inue explo ing he numbe ze o. 3.2.3. Educa ional dialogue o high sel -e icacy eache s In he class ooms o high sel -e icacy eache s, 43 episodes o educa ional dialogue we e iden i ied, and high-quali y eache -ini ia ed dialogues (n=24) p e ailed. The pos -hoc es using Dunn’s es showed ha he high sel -e icacy eache s u ilised signi ican ly mo e eache -ini ia ed dialogue o a high quali y compa ed wi h he low sel -e icacy eache s (z = -7.062, p =.043). A e eache -ini ia ed dialogues o a high quali y, he nex mos equen ly used dialogue ype was he child-ini ia ed dialogues o a high quali y (n = 10). The pos -hoc es using Dunn’s es showed ha he high sel -e icacy eache s used child-ini ia ed dialogue o a high quali y signi ican ly mo e compa ed wi h he mode a e sel -e icacy eache s (z = 7.000, p = .024). The mode a e-quali y dialogues occu ed less equen ly in he lessons o high sel -e icacy eache s: eigh episodes o eache -ini ia ed dialogue o a mode a e quali y and one episode o child-ini ia ed dialogue o a mode a e quali y we e ound. Example 3 p esen s an ex ac o a child-ini ia ed high-quali y dialogue, he pa e n ype ha was ound he second mos equen ly in he class ooms o high sel -e icacy eache s. The dialogue was conduc ed du ing an a lesson. S a ing om he e y beginning, he child en ac i ely pa icipa e in discussion and sha e hei hough s and expe iences (e.g., I ha e made animals ou o clay in p eschool; I ha e seen some hing in No way in a glass shop. Fi s , he e was only his lump, hen hey pu on a s ick and bu i on i e and hen hey could modi y and blow i .) The eache is open o he child en’s hough s and shows in e es in hem by p o iding eedback o asking u he ques ions (e.g., How did you know ha his is clay?; Yes, ha ’s glassblowing. Glass ac s a bi simila ly as clay in he kiln.). The child en also pa icipa e by asking hei own ques ions (How did hey make ha clay ha colou ?) and eplying o eache ’s ques ion (So ha i will ge ha d, almos like a s one.). Howe e , du ing he dialogue, he Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 19 eache asks only a ew ques ions and p edominan ly sha es in o ma ion o which he child en eac wi hou speci ic encou agemen (e.g., I ha e no iced ha qui e o en he clays ha we use in ou school come om England. Some clays a e also Finnish. ; Because i we lea e he ai bubbles he e he clay wo k will easily explode when we hea i .). The dialogic exchange is ecip ocal and cumula i e be ween he eache and child en. Example 3. Child-ini ia ed high-quali y dialogue o high sel -e icacy eache . Con ex : Teache b ings a pile o clay o he able o child en o explo e and make clay a . Child 1: Teache eache ! I ha e done clay a be o e. I ha e made animals ou o clay in p eschool. Child 2: Yes, bu hey we e g ey! Teache : How did you know ha his is clay? Child 3: You old us! Teache : Did I? I hink ha he umou s wen a ound al eady in he mo ning. Child en: Yeah!!!!! Teache : I ha e no iced ha qui e o en he clays ha we use in ou school come om England. I am no su e om which coun y his clay is om. Some clays a e also Finnish. Child 4: Yes! I could guess! Teache : And his clay has been al eady wo ked wi h machine. The machine has al eady aken ai bubbles ou o he clay. Child 5: And wo ms! Teache : Yes indeed his clay has been cleaned. The e is no hing ex a was e in i . Bu i you ook he clay s aigh om he g ound, you should wo k on i o a while o ge he ai bubbles ou . Because i we lea e he ai bubbles he e he clay wo k will easily explode when we hea i . Why do I wan o hea you clay wo ks a he end? Pu hem o a eally ho kiln. Child 6: I ha e seen some hing in No way in a glass shop. Fi s , he e was only his lump, hen hey pu on a s ick and bu i on i e and hen hey could modi y and blow i . Teache : Yes ha ’s glassblowing. Glass ac s a bi simila ly as clay in he kiln. Child 7: So ha i will ge ha d, almos like a s one. Teache : Yes, you a e igh . The kiln will solidi y and ha den he clay. Child 1: How did hey make ha clay ha colou ? Teache : This is ed clay. So his is he o iginal colou o his clay when i ’s on he g ound. The eache and child en con inue explo ing clay. Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 20 4. Discussion In he p esen s udy, we examined educa ional dialogue in he G ade 1 class ooms o eache s wi h low, mode a e and high sel -e icacy belie s. Episodes o educa ional dialogue we e ca ego ised among he ou pa e ns o dialogic eaching. In e ms o he occu ence o he dialogues, he esul s showed ha eache s wi h low sel -e icacy used educa ional dialogue he leas equen ly in hei class ooms. In e ms o he quali y o he dialogues, eache s wi h low sel -e icacy u ilised less eache -ini ia ed dialogue o a high quali y compa ed wi h mode a e and high sel -e icacy eache s. I was also ound ha al hough he mode a e and high sel -e icacy eache s conduc ed almos he same amoun o dialogic episodes in o al, he quali y o he dialogues in hei lessons a ied: high sel -e icacy eache s u ilised mo e child- ini ia ed dialogues o a high quali y compa ed wi h he mode a e sel -e icacy eache s. Fi s , we examined he occu ence o episodes o educa ional dialogue wi hin he class oom in e ac ions o low, mode a e and high sel -e icacy eache s. P e ious esea ch has shown ha eache s’ high sel -e icacy belie s os e hei wo k commi men (Klassen & Chiu, 2011) and can be linked wi h he educa ional p ac ices ha hey choose o use and in es e o and pe sis ence in (Bandu a, 1997; Wool olk Hoy e al., 2009). The indings o he p esen s udy showed ha low sel -e icacy eache s used educa ional dialogue he leas o en in hei class ooms, whe eas he mode a e and high sel -e icacy eache s used dialogue signi ican ly mo e. I is acknowledged he e ha conduc ing educa ional dialogue and sca olding s uden s’ lea ning h ough i can be challenging o eache s and equi es a high amoun o e o (Muhonen e al., 2018; So ensen & Takle, 2002). I is possible ha high and mode a e sel - e icacy eache s may eel mo e compe en and con iden in hei eaching and using educa ional dialogue, whe eas low sel -e icacy eache s may expe ience a lack o con idence o knowledge in ca ying ou and in es ing ime in dialogue. Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 21 I is also wo h conside ing ha eache s’ eaching expe iences may be linked wi h he eache s’ sel -e icacy and occu ence o educa ional dialogue. P io esea ch has shown ha eache s’ sense o e icacy ends o inc ease wi h yea s o eaching expe ience (Soodak & Podell, 1997). In he eache selec ion phase o he p esen s udy, i was in e es ing o no e ha eache s’ wo k expe ience a ied among he low, mode a e and high sel -e icacy eache s al hough he a e age sco e o eaching expe ience in he o al sample was ela i ely high (16.1 yea s). On a e age, he low sel -e icacy eache s we e ound o ha e he mos eaching expe ience (21.5 yea s), whe eas he mode a e sel -e icacy eache s had he leas eaching expe ience (13.7 yea s). Also he high sel -e icacy eache s had a high a e age eaching expe ience (19.5 yea s). Though he p edominan esea ch sugges s ha eache sel -e icacy inc eases wi h expe ience (Soodak & Podell, 1997), i has also been sugges ed ha he eaching expe ience may also ha e an e ec on how eache s pe cei e hei own abili ies and limi a ions and how hey hen can become mo e awa e o hese limi a ions (Egyed & Sho , 2006). The eache g oup o high sel -e icacy belie s and high eaching expe ience can be seen as ep esen ing he ‘ideal’ scena io in which he eache s’ sel -e icacy g ow h wi h yea s is e lec ed in hei high-quali y eaching p ac ices, in his case, in he use o educa ional dialogue. On he o he hand, he eache g oup o low sel -e icacy belie s and high eaching expe ience can be seen as ep esen ing a somewha ‘cynical’ scena io in which he eache s’ dec eased sel -e icacy o e he yea s and whe e low pe cep ions abou hei abili ies and limi a ions a e e lec ed in hei sca ce use o educa ional dialogue in he class oom. The less expe ienced eache s o he sample wi h mode a e sel -e icacy conduc ed educa ional dialogue he mos in hei class ooms. Recen ly, he impo ance o class oom in e ac ions and educa ional dialogue has been acknowledged in Finnish eache educa ion, and he e has been an endea ou o inc ease eache s uden s’ skills o conduc and sca old educa ional dialogue (Muhonen e al., 2016, 2018). Because hese eache s ecei ed hei eache educa ion mo e ecen ly, hei Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 22 highe use o educa ional dialogue may also e lec he quali y o he cu en eache educa ion in Finland. The second esea ch ques ion ocused on he quali y o educa ional dialogue by examining he pa e ns o dialogic eaching among low, mode a e and high sel -e icacy eache s. P e ious s udies (Muhonen e al., 2016, 2018, 2020) ha e shown ha eache -ini ia ed high-quali y dialogue is he p e alen dialogue pa e n in p ima y school and p eschool class ooms. The indings o he cu en s udy a e in line wi h p e ious s udies because eache - ini ia ed, high-quali y dialogues occu ed he mos equen ly in he cu en sample. Howe e , he indings showed signi ican a ia ion in he pa e ns o dialogic eaching be ween he h ee eache g oups. Mode a e and high sel -e icacy eache s u ilised eache -ini ia ed, high-quali y dialogue signi ican ly mo e compa ed wi h he low sel -e icacy eache s. Especially in he class ooms o mode a e sel -e icacy eache s, he clea majo i y o he iden i ied dialogic episodes ep esen ed eache -ini ia ed, high-quali y dialogue (33 episodes ou o 44 episodes). This indica es ha he mode a e and high sel -e icacy eache s may pu mo e e o in o sca olding dialogue and s uden s’ lea ning h ough i . The indings suppo p e ious esea ch, which sugges s ha eache s wi h highe sel -e icacy a e mo e likely o p o ide highe -quali y suppo o s uden s in hei class ooms, which can lead o a highe quali y in he eache – s uden class oom in e ac ions (Pe e a & John, 2020; Wool olk Hoy e al., 2009). Indeed, he mos equen ly ound eache -ini ia ed, high-quali y dialogues a e cha ac e ised by high- quali y eache suppo and sca olding. As Alexande (2000) sugges ed, in well-sca olded educa ional dialogue, he sha ed knowledge building is s uc u ed h ough ques ioning, join ac i i y and sha ed concep ions, which we e ound in he class ooms o mode a e and high sel - e icacy eache s. This ype o ac i e and e sa ile sca olding enables oppo uni ies o s uden pa icipa ion and concep ual de elopmen (Gillies, 2004: Gillies, 2015), bu on he o he hand, i is s ill p edominan ly led by he eache . Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 23 I is also impo an o acknowledge ha child-ini ia ed dialogues we e a e in he cu en sample o G ade 1 eache s and hei class ooms. Simila indings ha e been ound in esea ch on p ima y school and p eschool class ooms (Muhonen e al., 2016, 2018, 2020). Howe e , he indings o he cu en s udy added new knowledge by showing ha he high sel -e icacy eache s allowed and sca olded child-ini ia ed dialogues he mos ou o he h ee eache g oups. Al hough, in o al, he mode a e and high sel -e icacy eache s we e ound o ha e app oxima ely he same amoun o dialogic episodes, high sel -e icacy eache s conduc ed signi ican ly mo e child-ini ia ed, high-quali y dialogues compa ed wi h he mode a e sel - e icacy eache s. P io esea ch has shown ha eache s expe iencing high sel -e icacy ypically c ea e a mo e a ou able ins uc ional lea ning en i onmen o hei s uden s by u ilising mo e s uden -cen ed educa ional app oaches (Nie e al., 2013) and p o iding highe lea ning suppo (Holzbe ge e al., 2013). The esul s o he p esen s udy showed ha eache s expe iencing high sel -e icacy may eel mo e con iden abou hei educa ional abili ies and a e willing o in es mo e ime, space and e o o acili a e bene icial lea ning expe iences o s uden s h ough child-ini ia ed dialogues. Conduc ing child-ini ia ed dialogues can be demanding o eache s because hey canno necessa ily p edic whe e he dialogue may lead, and hey migh no ha e a p io plan o he discussion (Muhonen e al., 2016). In his ype o si ua ion, high sel -e icacy and us in one’s own p o essional knowledge may suppo he eache in allowing an open space o child-ini ia ed alk, being sensi i e and lexible du ing he discussion and s ill o ming and guiding a clea lea ning goal o he discussion. 4.1. Implica ions, u u e di ec ions, and limi a ions The p esen s udy has bo h heo e ical and p ac ical implica ions. Because p io esea ch has acknowledged he impo ance o educa ional dialogue in e ms o s uden s’ lea ning (e.g., Alexande , 2018; Muhonen e al., 2018; Sedo a e al., 2019), mo e e o should Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 30 and dialogue (pp. 333–347). Washing on, DC: Ame ican Educa ional Resea ch Associa ion. Muhonen, H., Paka inen, E., Le kkanen, M-K., Ba za, L., & on Suchodole z, A. (2020). Pa e ns o dialogic eaching in kinde ga en class ooms o Finland and he Uni ed A ab Emi a es. Lea ning, Cul u e and Social In e ac ion, 25, 100264. Doi:10.1016/j.lcsi.2018.11.011 Muhonen, H., Paka inen, E., Poikkeus, A.-M., Le kkanen, M.-K., & Rasku-Pu onen, H. (2018). Quali y o educa ional dialogue and associa ion wi h s uden s’ academic pe o mance. Lea ning and Ins uc ion, 55, 67–79. Doi:10.1016/j.lea nins uc.2017.09.007 Muhonen, H., Paka inen, E., Rasku-Pu onen, H., & Le kkanen, M.-K. (2020). Educa ional dialogue o eache s expe iencing di e en le els o s ess. Manusc ip submi ed o publica ion. Muhonen, H., Rasku-Pu onen, H., Paka inen, E., Poikkeus, A-M. & Le kkanen, M-K. (2016). Sca olding h ough dialogic eaching in ea ly school class ooms. Teaching and Teache Educa ion. 55, 143–154. Doi:10.1016/j. a e.2016.01.007 Myhill, D. (2006). Talk, alk, alk: Teaching and lea ning in whole class discou se. Resea ch Pape s in Educa ion, 21(1), 19–41. Doi:10.1080/02671520500445425 Nie, Y., Tan, G. H., Liau, A. K., Lau, S., & Chua, B. L. (2013). The oles o eache e icacy in ins uc ional inno a ion: I s p edic i e ela ions o cons uc i is and didac ic ins uc ion. Educa ional Resea ch o Policy and P ac ice, 12(1), 67–77. Doi:10.1007/s10671-012-9128-y Nys and, M. (1997). Opening dialogue. Unde s anding he dynamics o language and lea ning in he English class oom. New Yo k, NY: Teache s College P ess. Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 31 Pelle ie , D. (2006). Theo e ical conside a ions ela ed o cu o poin s. Food and Nu i ion Bulle in, 27, 224–236. Doi:10.1177/15648265060274S506 Pe e a, H. N., Calkins, C., & Pa , R. (2019). Teache sel -efficacy p ofiles: De e minan s, ou comes, and gene alizabili y ac oss eaching le el. Con empo a y Educa ional Psychology, 58, 186–203. Doi:10.1016/j.cedpsych.2019.02.006. Pe e a, H. N., & John, J. E. (2020). Teache s’ sel -e icacy belie s o eaching ma h: Rela ions wi h eache and s uden ou comes. Con empo a y Educa ional Psychology, 61, 101842. Doi:10.1016/j.cedpsych.2020.101842 Rasku-Pu onen, H., Le kkanen, M-K., Poikkeus, A.-M., & Siekkinen, M. (2012). Dialogical Pa e ns o In e ac ion in P eschool Class ooms. In e na ional Jou nal o Educa ional Resea ch, 53, 138–149. Doi: 10.1016/j.ije .2012.03.004 Rojas-D ummond, S., To eblanca, O., Ped aza, H., Vélez, M., & Guzmán, K. (2013). Dialogic sca olding: Enhancing lea ning and unde s anding in collabo a i e con ex s. Lea ning, Cul u e and Social In e ac ion, 2(1), 11–21. Doi:10.1016/j.lcsi.2012.12.003 Sedo a, K., Sedlacek, M., S a icek, R., Majcik, M., Na a ilo a, J., D exle o a A., … Salamouno a, Z. (2019). Do hose who alk mo e lea n mo e? The ela ionship be ween s uden class oom alk and s uden achie emen . Lea ning and Ins uc ion, 63, 101217. Doi:10.1016/j.lea nins uc.2019.101217 Skaal ik, E. M., & Skaal ik, S. (2007). Dimensions o eache sel -e icacy and ela ions wi h s ain ac o s, pe cei ed collec i e eache e icacy, and eache bu nou . Jou nal o Educa ional Psychology, 99, 611–625. Doi:10.1037/0022-0663.99.3. 611. Soodak, L. C., & Podell, D. M. (1997). E icacy and expe ience: pe cep ions o e icacy among p ese ice and p ac icing eache s. Jou nal o Resea ch and De elopmen in Educa ion 30(4), 214–21. Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 32 So ensen, E. K., & Takle, E. (2002). Collabo a i e knowledge building in web-based lea ning: Assessing he quali y o dialogue. In e na ional Jou nal on E-Lea ning, 1(1), 28–32. Re ie ed om h ps://lib.d .ias a e.edu/cgi/ iewcon en .cgi? e e e =h ps://www.- google.com/&h ps edi =1&a icle=1208&con ex =ge_a _pubs Tschannen-Mo an, M., & Wool olk Hoy, A. (2001). Teache e icacy: Cap u ing and elusi e cons uc . Teaching and Teache Educa ion, 17, 783-805. Doi: 10.1016/S0742- 051X(01)00036-1 Tschannen-Mo an, M., Wool olk Hoy, A., & Hoy, W. K. (1998). Teache efficacy: I s meaning and measu e. Re iew o Educa ional Resea ch, 68, 202–248. Doi:10.3102/00346543068002202. Tsouloupas, C. N., Ca son, R. L., Ma hews, R., G awi ch, M. J., & Ba be , L. K. (2010). Explo ing he associa ion be ween eache s’ pe cei ed s uden misbeha iou and emo ional exhaus ion: The impo ance o eache e icacy belie s and emo ion egula ion. Educa ional Psychology, 30(2), 173–189. Doi:10.1080/01443410903494460. Vygo sky, L. S. (1978). Mind in socie y: he de elopmen o highe psychological p ocesses. Camb idge: Ha a d Uni e si y P ess. Webb, N. M., F anke, M. L., De, T., Chan, A. G., F eund, D., Shein, P., & Melkonia, D. (2009). “Explain o you pa ne ”: Teache s’ ins uc ional p ac ices and s uden s’ dialogue in small g oups. Camb idge Jou nal o Educa ion 39(1), 49–70. Doi:10.1080/03057640802701986 Wells, G. (2009). Ins uc ional Con e sa ion in he Class oom: Can he Pa adox be Resol ed? Re ie ed om h ps://people.ucsc.edu/~gwells/Files/Pape s_Folde /documen s/ICAERA09.pd Running Head: EDUCATIONAL DIALOGUE AND TEACHER SELF-EFFICACY 33 Wilkinson, I. A. G., Rezni skaya, A., Bou dage, K., Oyle , J., Glina, M., D ew y, R., & Nelson, K. (2017). Towa d a mo e dialogic pedagogy: Changing eache s' belie s and p ac ices h ough p o essional de elopmen in language a s class ooms. Language and Educa ion, 31(1), 65–82. Doi:10.1080/09500782.2016. 1230129. Wool olk Hoy, A., Hoy, W. K., & Da is, H. A. (2009). Teache s' sel -e icacy belie s. In K. Wen zel, & A. Wig ield (Eds.). Handbook o mo i a ion a school (pp. 627–653). New Yo k, NY: Rou ledge. Zee, M., & Koomen, H. M. (2016). Teache sel -e icacy and i s e ec s on class oom p ocesses, s uden academic adjus men , and eache well-being: A syn hesis o 40yea s o esea ch. Re iew o Educa ional Resea ch, 86, 981–1015. Doi:10.3102/0034654315626801.